BVRI light curves for 22 Type Ia supernovae

Abstract

We present 1210 Johnson/Cousins B, V, R, and I photometric observations of 22 recent Type Ia supernovae (SNe Ia): SNe 1993ac, 1993ae, 1994M, 1994S, 1994T, 1994Q, 1994ae, 1995D, 1995E, 1995al, 1995ac, 1995ak, 1995bd, 1996C, 1996X, 1996Z, 1996ab, 1996ai, 1996bk, 1996bl, 1996bo, and 1996bv. Most of the photometry was obtained at the Fred Lawrence Whipple Observatory of the Harvard-Smithsonian Center for Astrophysics in a cooperative observing plan aimed at improving the database for SNe Ia. The redshifts of the sample range from cz = 1200 to 37,000 km s-1 with a mean of cz = 7000 km s-1.

Full text

arXiv:astro-ph/9810291v1 19 Oct 1998
BV RI Light Curves for 22 Type Ia Supernovae
Accepted to the Astronomical Journal
Adam G. Riess1, Robert P. Kirshner2,Brian P. Schmidt3,Saurabh Jha2, Peter Challis2, Peter M.
Garnavich2, Ann A. Esin2, Chris Carpenter2, Randy Grashius4, Rudolph E. Schild2, Perry L. Berlind5,
John P. Huchra2, Charles F. Prosser6, Emilio E. Falco2, Priscilla J. Benson7, Cesar Briceno2, Warren R.
Brown2, Nelson Caldwell5, Ian P Dell’Antonio8, Alexei V. Filippenko1, Alyssa A. Goodman2, Norman A.
Grogin2, Ted Groner5, John P. Hughes9, Paul J. Green2, Rolf A. Jansen2, Jan T. Kleyna2, Jane X. Luu2,
Lucas M. Macri2, Brian A. McLeod2, Kim K. McLeod7,Brian R. McNamara2, Brian McLean10 , Alejandra
A. E. Milone11,Joseph J Mohr12 , Dan Moraru2, Chien Peng1 13, Jim Peters5, Andrea H. Prestwich2,
Krzysztof Z. Stanek2, Ping Zhao2
ABSTRACT
We present 1210 Johnson/Cousins B,V,R, and I photometric observations of 22 recent type
Ia supernovae (SNe Ia): SN 1993ac, SN 1993ae, SN 1994M, SN 1994S, SN 1994T, SN 1994Q,
SN 1994ae, SN 1995D, SN 1995E, SN 1995al, SN 1995ac, SN 1995ak, SN 1995bd, SN 1996C, SN
1996X, SN 1996Z, SN 1996ab, SN 1996ai, SN 1996bk, SN 1996bl, SN 1996bo, and SN 1996bv.
1Department of Astronomy, University of California, Berkeley, CA 94720-3411
2Harvard-Smithsonian Center for Astrophysics, 60 Garden St., Cambridge, MA 02138
3Mount Stromlo and Siding Spring Observatories, Private Bag, Weston Creek P.O. 2611, Australia
4University of New Mexico, Capilla Peak Observatory Abuquerque, NM 87131
5Fred Lawrence Whipple Observatory, Amado, AZ 85645
6NOAO, Tucson, AZ 85726
7Whitin Observatory, Wellesley College, MA 02481-8203
8Bell Lab, Murray Hill, NJ 07974
9Rutgers University, Dept. of Physics & Astronomy, New Brunswick, NJ 08855
10Space Telescope Science Institute, Baltimore, MD 21218
11Multiple Mirror Telescope Observatory, c/o Whipple Observatory,P.O. Box 97, Amado AZ 85645-0097
12Department of Astronomy and Astrophysics, University of Chicago, Chicago, IL 60637
13Steward Observatory, University of Arizona, Tucson, AZ 85721

–2–
Most of the photometry was obtained at the Fred Lawrence Whipple Observatory (FLWO)
of the Harvard-Smithsonian Center for Astrophysics in a cooperative observing plan aimed at
improving the data base for SN Ia. The redshifts of the sample range from cz=1200 to 37000
km s−1with a mean of cz=7000 km s−1.

–3–
1. Introduction
Recent evidence suggests that type Ia supernovae (SNe Ia) can be used as exceedingly precise long-range
distance indicators (Riess, Press, & Kirshner 1995a,b, 1996a,b; Hamuy et al. 1995,1996a,b; Maza et al.
1994; Phillips 1993; Tammann & Sandage 1995). With peak luminosities a million times greater than
Cepheid variables and individual distance accuracy approaching 5%, they provide cosmology with a tool of
great leverage.
The uses for these extragalactic beacons are numerous. As test particles in the nearby Hubble flow,
they have been used to measure the current expansion rate of the Universe (Sandage et al. 1992, 1994,
1996; Sandage & Tammann 1993; Schaefer 1994, 1995a,b, 1996; Branch & Tammann 1992; Tammann
& Leibundgut 1990; Arnett, Branch, & Wheeler 1985; Cadonau, Sandage, & Tammann 1985; Hamuy et
al. 1995; 1996a,b; Riess, Press, & Kirshner 1995a, 1996a; Riess et al. 1998a; Tripp 1998; Branch 1998
and references within). Combined with their positions on the sky, SNe Ia have been used to reveal the
peculiar velocities of distant galaxies as well as the bulk flow of our own local neighborhood (Tammann &
Leibundgut 1990; Miller & Branch 1992; Jerjen & Tammann 1993; Riess, Press, & Kirshner 1995b; Watkins
& Feldman 1995; Riess et al. 1997b; Zehavi et al. 1998; Tammann 1998). SN Ia evolution is the only
well understood time-variable process which can be used to mark the passage of time at high redshift. As
cosmological clocks, SNe Ia have been used to examine the nature of the redshift using the time dilation test
(Rust 1974; Leibundgut 1990; Goldhaber et al. 1997; Leibundgut et al. 1996; Riess et al. 1997a). SNe Ia
have been employed as probes of extragalactic dust (Della Valle & Panagia 1993; Riess, Press, & Kirshner
1996b), and their contribution to galactic chemical enrichment by their production of iron peak elements
has been explored by measuring their rates of occurrence (Timmes 1991; Cappellaro et al. 1993a,b, 1997;
Turatto et al. 1994; van den Bergh & McClure 1994; Pain et al. 1997; Madau, Della Valle, & Panagia
1998). Recently, vigorous programs have embarked on searches for SNe Ia at high redshifts (0.2 ≤z≤
1.0) with the intent of measuring the expansion history of the Universe (Perlmutter et al. 1995, 1997;
Schmidt et al. 1998). Early results imply that there is not enough gravitating matter to close the Universe
(Garnavich et al. 1998a; Perlmutter et al. 1998) and that currently the expansion is accelerating (Riess
et al. 1998b; Perlmutter 1999). Supernova observations, when combined with measurements of Cosmic

–4–
Microwave Background anisotropies, may prove useful to determine the cosmic equation of state and the
global geometry of the Universe (Garnavich et al. 1998b; White 1998).
These applications require well-observed SN Ia light curves with reliable photometry. Further, the
cosmological applications rely on comparisons to nearby SNe Ia which delineate today’s Hubble flow
(0.01 < z < 0.1).
Despite the importance of precisely observed SN Ia light curves, most published SN Ia photometry,
before 1980, consisted of infrequently sampled photographic light curves for SNe within cz < 2000 km s−1
with a wide assortment of filters and emulsions (van den Bergh 1994; Cadonau & Leibundgut 1990; Barbon,
Cappellaro, & Turatto 1989). Much of this data is plagued by systematic errors from non-linearities in
detector sensitivity, uncertain transformations to modern filter conventions, and difficulties with background
light subtraction. A crude estimate of these photometry errors comes from comparing the dispersion in
Sandage & Tammann’s (1993) Hubble diagrams constructed with old photographic light curves (0.65 mag)
with those based on more modern B-band SN Ia observations (0.38 mag; Hamuy et al. 1996a). The result
suggests that typical errors from the pre-1980 photographic SN Ia photometry could be as high as ∼0.5
magnitudes.
Obtaining well sampled light curves with high-precision photometry (σ≤0.03 mag) is challenging.
Collecting observations of SNe Ia in each of four filters every few days for ∼100 days is a task that is
not well suited to the short blocks of time allocated at most modern observatories. Variable weather
poses another challenge to obtaining well-sampled SN Ia light curves. It is also important to maintain a
telescope-detector setup throughout the observations which well matches standard passband conventions
(Johnson & Harris 1954; Bessell 1990) since the non-stellar spectrum of an SN Ia can make linear color
corrections inexact. The most challenging obstacle to producing a high-quality SN Ia light curve is to
account correctly for the background light from the host galaxy at the position of the supernova.
Recent evidence has shown that SNe Ia are not perfectly homogeneous in luminosity or color and
that the intrinsic luminosity and color is intimately related to the shape of the observed light curves
(Phillips 1993; Riess, Press, & Kirshner 1995a, 1996a; Riess et al. 1998b; Hamuy et al. 1995, 1996a).
Incorrectly subtracting the background light from a set of supernova observations can have disastrous

–5–
effects on the light curve shape. Boisseau & Wheeler (1991) have investigated the effect of background
galaxy contamination on the inferred absolute magnitude and light-curve speed of SNe Ia. They find that
oversubtraction or undersubtraction of a constant flux source leads to an observed correlation between SN
Ia light curve speed and inferred luminosity. If the value of this correlation were the same as the intrinsic
correlation, then using the light-curve shape to correct the luminosity could equally well account for either
intrinsic luminosity variation or galaxy contamination. Unfortunately, they are not the same, so it is crucial
to account correctly for the background light to determine the true light-curve shape. SN Ia light curves
obtained by the Cal´an/Tololo survey show that with high quality photometry, it is possible to measure
distances with SN Ia light curves to a precision approaching ∼5% (Hamuy et al. 1995, 1996a; Riess, Press
& Kirshner 1995a, 1996a).
Since the widespread use of modern CCD detectors coupled with commercially available
Johnson/Cousins passbands began in ∼1980, there have been nearly 250 SNe Ia reported (van den
Bergh 1994; Barbon, Cappellaro, & Turatto 1989). Regrettably, light curves have been collected, reduced,
and published for fewer than 50 of these objects (Cadonau & Leibundgut 1990; Hamuy et al. 1993; Hamuy
et al. 1994; 1996b; Sadakane et al. 1996; Phillips 1993 and references within). Of these, less than half
include Iand Rlight curves which, combined with shorter wavelength light curves, can be used to determine
the reddening due to dust (Ford et al. 1993; Hamuy et al. 1996b).
Recently, progress has been made toward building a reliable sample of SN Ia light curves in the Hubble
flow. The largest contribution to date has been made by the Cal´an/Tololo Supernova Survey, a program
begun in 1990 by astronomers at Cerro Tololo Inter-American Observatory (CTIO) and the Cerro Cal´an
Observatory of the University of Chile (Hamuy et al. 1993). This photographic search with follow-up B,V,I
CCD photometry netted 27 SNe Ia with (0.01 < z < 0.1). Other programs which show great promise
include the Beijing Astronomical Observatory search (IAUC 6379), the Mount Stromlo Abell Cluster
Supernova Search (Reiss et al. 1998), and the Lick Observatory Supernova Search (IAUC 6627), all of
which have made repeated discoveries and in the future are expected to contribute to the growing sample
of SN Ia light curves.
Beginning in 1993, astronomers at the Center for Astrophysics (CfA) began a concerted and organized

–6–
effort to collect Johnson/Cousins BV RI photometry of type Ia supernovae. Many of these SNe Ia were
discovered serendipitously by amateurs or by professionals scanning images collected for another purpose.
This work presents the light curves of 22 of SNe Ia in the Hubble flow observed between 1993 and 1996. In
§2 we give details of our observational setup and reduction procedure. We present BV RI photometry for
22 SNe Ia in §3. In §4 we discuss the characteristics of this sample.
2. Observations & Reductions
Most of the photometric data presented in this paper were collected at the 1.2 m telescope at the Fred
Lawrence Whipple Observatory (FLWO). The 1.2 m is an f /8 Ritchey-Chretien reflector and was outfitted
with a thick front-illuminated Loral CCD between 1993 and July 1995, and a thinned, back-side illuminated
Loral CCD from August 1995 through 1996. The surfaces of both CCDs were coated with a laser dye which
improves the blue sensitivity. The pixels are 15 microns square, corresponding to 0.31 arcseconds at the
focal plane of the 1.2 m telescope. The filters are constructed from Schott glass components recommended
by Bessell (1990) for a coated CCD.
The B, V , R, and ICCD transmission functions for the FLWO 1.2 m telescope are shown in Figure 1
(Andy Szentgyorgyi, private communication). The CCD’s ability to detect light over a range of wavelengths
makes it difficult to emulate the sharp blue-side or red-side cut-offs of photomultiplier transmission
functions. The Band VCCD transmissions closely match the photomultiplier passband conventions of
Johnson & Harris (1954) and most recently Bessell (1990). The R-band CCD transmission is similar to
Cousins (1980, 1981) and Bessell (1990). The I-band CCD transmission extends to substantially longer
wavelengths than the Cousins (1980, 1981) and Bessell (1990) convention for the Iphotomultiplier passband.
The FLWO CCD transmissions are very similar to CCD transmission functions obtained by Bessell with
the filters he prescribed (Bessell 1990). We use linear color corrections to account for differences between
our CCD transmission functions and the Johnson/Cousins photomultiplier passbands. Still, broad emission
and absorption features in the spectral energy distribution of SN Ia can cause variations among light
curves observed with slightly different CCD transmission functions. The difference between broad-band
SN Ia photometry obtained at FLWO and CTIO has been determined in detail by Smith et al. (1998) by
comparing phototmetry of SN 1994D obtained at the two sites. These differences (with uncertainties in

–7–
parenthesis), as shown in Table 1, are small, but not entirely absent. Agreement between FLWO and CTIO
is best in the Vpassband.
A small number of the observations reported here were conducted at the 0.62 m telescope at the
Capilla Peak Observatory (CPO). The 0.62 m is a f/15.2 Boller & Chivens Cassegrain telescope matched
with a RCA model SID501EX back-illuminated and thinned CCD. The CCD pixels are 30 microns square,
corresponding to 0.67 arcseconds per pixel at prime focus. The CPO CCD transmission functions are shown
by Beckert & Newberry (1989) and are a good match to the CCD transmissions at FLWO. The mean
difference in SN Ia photometry obtained at FLWO and CPO is shown in Table 1 using SN 1995D. The
difference is very small in Vand somewhat larger in B, R, and I. Approximately 10 of the more than 1200
photometric observations were collected at other observatories including the Michigan-Dartmouth-MIT
Observatory, the McDonald Observatory, the McGraw Hill Observatory, the Lick Observatory, and CTIO
as noted in the photometry tables.
A well-sampled SN Ia light curve requires monitoring every day or two near maximum light and every
few days as it changes more slowly a fortnight after maximum. Weather and moonlight often intercede even
when an observatory is well-organized and well-instrumented to gather these observations. Fortunately,
photometric weather is not required since we can use stars in the SN field as local calibrators to monitor
the changing brightness of the SN Ia through differential photometry. Optimal comparison stars are the
brightest stars in the field which do not saturate the CCD electron wells during the long SN Ia exposures
at late times when the supernova has dimmed. Comparison stars obviate the need to make first-order
airmass corrections and allow the SN Ia brightness to be measured in any weather conditions for which the
supernova is visible.
On nights which are photometric, we performed all-sky photometry from which we constructed a
transformation from our detector measurements to the standard photometric conventions. Following Hardie
(1962) and Harris et al. (1981), we used transformation equations which gave the apparent magnitude as
a linear combination of the instrumental magnitude, observed airmass, and color. Using all-sky standard
stars from Landolt (1992) we then solved for the linear coefficients of the transformation equations. The
typical rms scatter of our transformations was 0.02 mag, with no observed correlation between residuals

–8–
and color, airmass, or instrumental magnitude. The mean color terms for the FLWO 1.2 m were 0.04, -0.03,
-0.08, and 0.06 mag in B , V, R, and Iper mag of B−V,B−V,V−R, and V−I, respectively. These
transformations were employed to calibrate the apparent magnitudes of the comparison stars which were
observed on the same night as the Landolt standards.
The comparison stars used for each SN Ia field are marked in Figure 2 and their B, V, R, and Iapparent
magnitudes are given in Table 2. For two fields (as listed in Table 2), only one “primary” comparison star
was consistently visible in the field of view.
For each night a SN Ia was observed, we measured the brightness of the supernova relative to one or
more comparison stars in the field. Extreme care was exercised to subtract properly the background light
at the location of the supernova. For SNe Ia far from the galaxy, or on a smooth and uniform region of the
galaxy, reliable background subtraction was easily accomplished. In such cases, we generally measured the
background light from the median sky value contained in an annulus of width ∼6′′ at an inner distance of
∼8′′ from the center of the supernova. These separations were increased as needed for data from nights
with poor seeing. More challenging were SNe Ia located on a luminous and mottled galaxy background. In
this case, unless images of the galaxy existed prior to explosion, the only reliable way to proceed was to
wait for the SN Ia to fade away to obtain an image of the galaxy without the supernova. Then by carefully
matching the alignment, intensity, and point-spread functions of the images with and without the SN Ia
present, we subtracted the two images to obtain an image of the supernova with zero background (Schmidt
et al. 1998). This is the same method used for the photometry of high-redshift SNe Ia (Riess et al. 1998b).
To measure the brightness of the supernova relative to the comparison stars in an uncrowded field we
used the method of aperture photometry. We added the light contained in a series of apertures of increasing
radius around the star and supernova and found the difference in magnitude between the SN Ia and the
comparison star which was independent of aperture radius. The estimated background was varied until a
difference in magnitude was found which was independent of aperture radius. This procedure is refined for
crowded fields or faint SNe Ia where we have fit point spread functions (PSFs) to the SN Ia and comparison
stars to determine their relative brightness. Experience has shown that when either technique is suitable,
the aperture method and the PSF method give consistent results within 0.01 mag. The particular method

–9–
used to derived the photometric measurements for each SN is listed in Table 3.
3. SN Ia Data
B, V , R, and Iband photometry for 22 SN Ia light curves is given in Tables 6 through 27 and plotted
in Figure 3. For each observation we include an estimate of the 1 σerror which was determined from
Poisson statistics, image quality, and uncertainty in the calibration of the comparison stars. In most cases
the dominant source of uncertainty is the comparison star calibration. In Table 3 we give details relevant
to the SN Ia observations including heliocentric redshift (column 2), the peak of the Blight curve (column
2), the peak of the Vlight curve (column 3), the decline in B15 days after maximum, ∆m15 (B) (column
4), time of the first observation relative to Bmaximum (column 5) as determined by the multicolor light
curve shape (MLCS) fit, right ascension (column 6), declination (column 7), and the photometric reduction
technique (column 8) used to measure the SN Ia’s brightness. The redshifts for SNe 1994S, 1994ae, 1995D,
1995E, 1995al, 1996Z, 1996ai, 1996bo, 1996bk, and 1996bv are from Huchtmeier & Richter (1989); SN
1993ae is from Chincarini & Rood (1977); SN 1995ak is from IAUC 6254; SN 1996bo is from IAUC 6492;
and SN 1996X is from the RC3 catalogue (de Vaucouleurs et al 1991). The rest were determined from our
spectra of the host galaxies.
The peaks of the Band Vlight curves and values of ∆m15(B) were determined from a light curve
fitting method which was different from that employed by Hamuy et al. (1996b). These values should not
be compared directly to the values given by Hamuy et al. (1996b). The only consistent way to combine the
parameters of the SNe Ia here and in Hamuy et al. (1996b) is to fit all of these data with a single fitting
method.
Table 4 lists information relevant to the host galaxies of the CfA SN Ia sample. This includes the
galaxy designation (column 2), the morphological type (column 3), the B−Vcolor of the host galaxy,
and the offset between the galaxy center and the SN Ia. The offsets were determined from a flux weighted
centroid for the SN and the galaxy in V. The B−Vgalaxy colors were determined from the largest
apertures (typically ∼20”) which avoided any foreground point source contamination. The same size
aperture was used to measure the galaxies magnitudes in Band V. The measured galaxy colors have an

– 10 –
uncertainty of σ= 0.1 mag.
Table 5 contains information relevant to the discovery and identification of each SN Ia. This includes
the discoverer and IAUC announcement (column 2), the method of image recording (column 3), the date of
discovery (column 4), and the observers who provided the spectral classification of the supernova (column
5). We have obtained spectra of each of these SNe Ia from the FLWO and after thorough examinination we
have found that each was of type Ia as defined by Branch, Fisher, & Nugent (1993) and Filippenko (1997).
Two objects, SN 1995ac and SN 1995bd, displayed similar spectral characteristics as the peculiar SN Ia
1991T including strong Fe III and weak Si II absorption (Garnavich et al. 1996).
4. Discussion
Many of these SNe Ia have been previously utilized for a variety of cosmological measurements as
discussed in §1. Nearly all of these applications make use of estimates of the luminosity distances to these
SNe Ia. The discussion of the most precise way to infer these distances has evolved from the assumption
of homogeneity (Leibundgut 1988; Branch & Miller 1993; Sandage & Tammann 1995) to methods which
account for the correlation between light-curve shape and luminosity (Riess, Press, & Kirshner 1995a;
Hamuy et al. 1995,1996a) and employ multiple passbands to separate the effects of dust on SN Ia light from
those of luminosity (Riess, Press, & Kirshner 1996a; Riess et al. 1998b). These methods are continually
evolving and improving and the specific values of the distance related parameters will likely be superseded
regularly. For this reason, we explore characteristics of this sample which are largely independent of the SN
distances.
In Figure 4 we show histograms of the supernova redshifts, peak apparent magnitude, epoch of first
observation, number of observations, absolute magnitude determined from the luminosity/light-curve
parameter, and line-of-sight extinction. For comparison we include the same statistics for the 27 SNe Ia
from the Cal´an/Tololo Supernova Survey (C/T). The CfA sample has a redshift range of 0.003 < z < 0.124
with a mean redshift of z= 0.025. As seen in Figure 4, the redshifts are concentrated at lower values,
though most are within the Hubble flow: in the rest frame of the cosmic microwave background (CMB), 17
of the 22 SNe Ia have cz > 2500 km s−1. For the C/T sample the redshift range is 0.011 < z < 0.101 with

– 11 –
a mean of z= 0.045.
The following sample characteristics are derived from multi-color light curve shape (MLCS) fits to
the BV RI data as described by Riess, Press, & Kirshner (1996a) and reanalyzed by Riess et al. (1998a);
they are subject to future refinements of fitting methods. The fitted peak apparent magnitudes for the
CfA sample range from 13.16 < mV<19.52 with a mean of mV= 15.70 ±1.65. For the C/T sample
the range is 14.64 < mV<19.35 with a mean of mV= 17.24 ±1.30. The epoch of the first light curve
observation for the CfA sample ranges from 12 days before maximum to 10 days after maximum. The
average starting epoch is coincident with Bmaximum with half of the SNe beginning before this time.
The C/T sample ranges between 10 days before to 12 days after maximum with a third beginning before
maximum. Figure 4 shows a histogram of the absolute magnitudes as inferred from the MLCS light-curve
shape fits on the Cepheid distance scale (Riess, Press, & Kirshner 1996a). The range of luminosities implied
for the CfA sample is −19.87 < MV<−18.80 with a mean of MV=−19.40 ±0.28; the C/T sample has
−19.68 < MV<−18.81 with a mean of MV=−19.27 ±0.29. A true SN Ia luminosity function can only
be derived from a sample of SNe Ia with well understood selection criteria (Reiss et al. 1998). Figure 4 also
shows the distribution of visual band extinctions as inferred from the MLCS measurements of reddening.
This distribution is strongly peaked toward low extinctions, with three notable exceptions (SN 1995E, SN
1995bd, and SN 1996ai) each having more than one visual magnitude of obscuration. One of these, SN
1995bd, is expected to have 1.5 mag of visual extinction from the Milky Way Galaxy (Schlegel, Finkbeiner,
& Davis 1998). Four more objects (SNe 1993ac, 1996bv, 1996bo, and 1996bk) have 0.5-1.5 mag of visual
extinction.
While the completeness and biases of the CfA SN Ia sample are not easily defined, it is still interesting
to combine these supernovae with the C/T sample to look for patterns in a large data set. Figure 5 shows
the extinction and light-curve shape parameters as a function of supernova galactocentric distance and
host galaxy type. Host galaxy extinction decays rapidly with projected separation from the nucleus and
with the progression to earlier type galaxies. The multicolor light-curve shape parameter (Riess, Press, &
Kirshner 1996a) shows that slow decline rates (∆ <0) dominate at small galactocentric distances and that
there is a general trend, first pointed out by Hamuy et al. (1996), for faster decline rates for supernovae

– 12 –
occurring in early-type galaxies. Figure 6 shows the distribution of absolute magnitude for SNe Ia in the
Hubble flow versus projected separation from the host galaxy. When no correction is made for extinction
or light-curve shape, the luminosity variation is similar to that found by Wang, H¨oflich, & Wheeler (1997);
there is a large dispersion at small galactocentric distances which decreases outward. However, when the
luminosity is corrected for total extinction (Figure 6b) from MLCS, SNe with projected separations of less
than 10 kpc are found, on average, to be brighter by about 0.3 mag than those further out. Because the
projected separation is the minimum distance the supernova can be from the galaxy center, the few faint
objects at small projected separations could be at even larger galactocentric distances. Elliptical hosts
dominate the sample at large separations, so the decrease of SN Ia luminosity in early-type hosts found by
Hamuy et al. (1996a) may contribute to this trend. When the luminosity is corrected for both extinction
and light-curve shape, no trend with galactocentric distance is apparent.
A Hubble diagram of the 17 SNe Ia from the CfA sample with cz > 2500 km s−1is given in Figure
7. These distances were derived with the MLCS method (Riess, Press, & Kirshner 1996a) as prescribed
by Riess et al. (1998b) in B, V , R, and I. The dispersion of these distances is σ= 0.16 mag. As noted by
Zehavi et al. (1998), the SNe Ia within cz ≈7000 km s −1(logcz ≈3.85) exhibit the dynamic complement
of a local void: an increased expansion rate relative to the more distant SNe Ia (see also Tammann 1998).
For these SNe Ia, the difference between the expansion rates within and beyond ∼7000 km s−1is 7%. It is
interesting to note that the sense of this change in the expansion rate is opposite to what would be caused
by a selection bias that emphasizes more luminous supernovae at larger distances. The observed effect
corresponds to a higher Hubble constant inferred locally. If a more statistically significant sample of SNe
Ia upholds this hint of a local void, it would help explain why SNe Ia yield a lower Hubble constant than
distance indicators that refer to more local volumes (Freedman et al. 1998; Jacoby et al. 1992).
The 22 BV RI SN Ia light curves presented here are composed of 1210 individual observations, and
comprise some of the most highly sampled illustrations of the photometric history of SNe Ia in the Hubble
flow. A histogram of the number of observations for each SN Ia is shown in Figure 4. Standouts include
SNe 1994ae, 1995D, 1995al, 1995bd, 1995ac, and 1996X, each with 80 to 100 observations beginning before
maximum and extending 60 to 100 days after maximum. Figure 8 shows BV RI composite light curves

– 13 –
formed by normalizing the data in time and brightness at the fit to the initial peak, including a correction
for 1 + ztime dilation and a K-correction (Hamuy et al. 1993). By affixing the light curves at maximum,
their inhomogeneity is readily apparent. For example, in Bthe light curves exhibit a decline in B15 days
after maximum [∆m15(B)] of 0.85 <∆m15 (B)<1.55 mag and in Ithe time and brightness of the second
maximum exhibits considerable variations. These and other features of the light curves have been shown
to correlate with the luminosity and colors of SNe Ia (Phillips 1993; Riess, Press, & Kirshner 1995a, 1996a;
Hamuy et al. 1995, 1996a,b). It is this property of SNe Ia which has recently enhanced their precision
as distance indicators and may lead to a better understanding of their progenitors and physical structure
(H¨oflich, Wheeler & Thielemann 1998).
We are deeply indebted to numerous observers on the 1.2 meter at Mt. Hopkins who graciously
observed our supernovae and enabled us to construct light curves of unprecedented sampling. We also wish
to thank Paul Schechter and Denise Hurley who contributed to the compilation of these data. The work at
U.C. Berkeley was supported by the Miller Institute for Basic Research in Science as well as NSF grant
AST-9417213, Supernova research at Harvard is supported by the NSF through grants AST-9528899 and
AST-9218475.

– 14 –
References
Arnett, W.D.,Branch, D., & Wheeler, J.C. 1985, Nature 314, 337
Barbon, R., Cappellaro, E., & Turatto, M. 1989, A&AS, 81, 421
Beckert D.C. & Newberry, M.V. 1989, PASP, 101, 849
Bessell, M.S. 1990, PASP, 102, 1181
Boisseau, J.R., & Wheeler, J.C. 1991, AJ, 101, 1281
Branch, D. 1998, ARAA, in press
Branch, D., & Miller, D. 1993, ApJ, 405, L5
Branch, D. & Tammann, G.A. 1992, ARA&A, 30, 359
Branch, D., Fisher, A., & Nugent, P. 1993, AJ, 106, 2383
Cadonau, R., Sandage, A., & Tammann, G.A. 1984, Lecture Notes in Physics, V224, 151
Cadonau, R., & Leibundgut, B. 1990, A&AS, 82, 145
Cappellaro, E., Turatto M., Bennetti, S., Tsvetkov, D. Yu., Bartunov, O. S., & Makarova, I. J. 1993a,
A&A, 268, 472
Cappellaro, E., Turatto M., Bennetti, S., Tsvetkov, D. Yu, Bartunov, O. S., & Makarova, I. J. 1993b, A&A,
283, 383
Cappellaro, E. et al. 1997, A&A, 322, 431
Chincarini, G., & Rood, H. 1977, ApJ, 214, 351
Cousins, A. W. J. 1980, S. Afr. Astron. Obs. Circ, 1, 166
Cousins, A. W. J. 1981, S. Afr. Astron. Obs. Circ, 6, 4
de Vaucouleurs, G. et al. 1991, in Third Reference Catalogue of Bright Galaxies (Springer-Verlag, New
York)
Della Valle, M., & Panagia, N. 1992, AJ, 104,696
Ford, C. et al. 1993, AJ, 106, 3
Freedman, W. et al. 1998, astro-ph/9801080
Garnavich, P. M., et al. 1998a, ApJ, 493, 53
Garnavich, P. M., et al. 1998b, ApJ, in press

– 15 –
Garnavich, P. M., et al. 1996, AAS, 189, 4509
Goldhaber, G., et al., 1997, in Thermonuclear Supernovae, ed. P. Ruiz-Lapuente, R. Canal, & J. Isern,
Dordrecht: Kluwer, p. 777
Hamuy, M., Phillips, M. M., Suntzeff, N. B., Schommer, R. A., Maza, J., & Avil´es, R. 1996a, AJ, 112, 2398
Hamuy, M., et al. 1996b, AJ, 112, 2408
Hamuy, M., Phillips, M. M., Maza, J., Suntzeff, N. B., Schommer, R. A., & Aviles, A. 1995, AJ, 109, 1
Hamuy, M., Phillips, M. M., Maza, J., Suntzeff, N. B., Schommer, R. A., & Aviles, A. 1994, AJ, 108, 2226
Hamuy, M., et al. 1993, AJ, 106, 2392
Hardie, R. H., 1962, in Stars and Stellar Systems, Vol. 2, Astronomical Techniques, ed. W.A. Hiltner,
(Chicago: University of Chicago Press), p. 198
Harris, W. E., Fitzgerald, M. P., & Reed, B. C. 1981, PASP, 93, 507
Hatano, K., Branch, D., & Deaton, J. 1998, ApJ, 502, 177
H¨oflich, P., Wheeler, J. C., & Thielemann, F. K. 1998, ApJ, 495, 617
Huchtmeier, W., & Richter, G., 1989, in A General Catalog of HI Observations of Galaxies, (Berlin:
Springer)
Jacoby, G. H., et al. 1992, PASP, 104, 599
Johnson, H. L., & Harris, D. L., 1954, ApJ, 120, 196
Jerjen, H., & Tammann, G.A. 1993, A&A, 276, 1
Landolt, A.U. 1992, AJ, 104, 340
Leibundgut, B. et al. 1996, ApJ, 466, L21
Leibundgut, B. 1990, A&A, 229, 1
Madau, P., Della Valle, M., & Panagia, N., 1998, MNRAS, 297,17
Maza, J., Hamuy, M., Phillips, M., Suntzeff, N., & Aviles, R. 1994, ApJ, 424, L107
Nugent, P., Phillips, M., Baron, E., Branch, D., & Hauschildt, P., 1995, ApJ, 455, L147
Pain, R. et al., 1997, in Thermonuclear Supernovae, ed. P. Ruiz-Lapuente, R. Canal, & J. Isern, Dordrecht:
Kluwer, p. 790
Perlmutter, S., et al., 1998, Nature, 391, 51
Perlmutter, S., et al., 1997, ApJ, 483, 565

– 16 –
Perlmutter, S. et al., 1995, ApJ, 440, 41
Phillips, M. 1993, ApJ, L105
Riess, A. G., Nugent, P. E., Filippenko, A. V., Kirshner, R. P., & Perlmutter, S., 1998a, ApJ, 504, 935
Riess, A. G. et al., 1998b, AJ, 116, 1009
Riess, A. G., et al. 1997a, AJ, 114, 722
Riess, A. G., Davis, M., Baker, J., & Kirshner, R. P. 1997b, ApJ, 488, L1
Riess, A. G., Press W. H., & Kirshner, R. P. 1996a, ApJ, 473, 88
Riess, A. G., Press W. H., & Kirshner, R. P. 1996b, ApJ, 473, 588
Riess, A. G., Press W. H., & Kirshner, R. P. 1995a, ApJ, 438, L17
Riess, A. G., Press W. H., & Kirshner, R. P. 1995b, ApJ, 445, L91
Rust, B.W., PhD Thesis, Oak Ridge National Laboratory (ORNL-4953)
Sadakane, K. et al., 1996, PASJ, 48, 51
Saha, A. et aly, 1994, ApJ, 425, 14
Sandage, A. et al. 1996, ApJ, 460, L15
Sandage, A. et al. 1994, ApJ, 423, L13
Sandage, A., & Tammann, G. A., 1993, ApJ, 415, 1
Sandage, A., Tammann, G. A., Panagia, N., & Macchetto, D. 1992, ApJ, 401, L7
Schaefer, B. E., 1996, ApJ, 464, 404
Schaefer, B. E., 1995a, ApJ, 447, L13
Schaefer, B. E., 1995b, ApJ, 449, L9
Schaefer, B. E., 1994, ApJ, 426, 493
Schlegel, D. J., Finkbeiner, D. P., & Davis, M. 1998, ApJ, 500, 525
Schmidt, B. P., et al. 1998, ApJ, in press
Shaw, R. L. 1979, A&A, 76, 188
Smith, C. et al. 1998, in preparation
Strauss, M. A., & Willick, J. A. 1995, PhR, 261, 271
Tammann, G. A., 1998, astro-ph/9805013
Tammann, G. A., & Sandage, A. 1995, ApJ, 452, 16

– 17 –
Tammann, G. A., & Leibundgut, B. 1990, A&A, 236, 9
Timmes, F. X., 1991, in Supernovae, ed S.E. Woosley, Springer Verlag, New York
Tripp, R., A&A, 1998, 331, 815
Turatto, M., Cappellaro, E., & Benetti, S. 1994, AJ, 108, 202
van den Bergh, S., & McClure, R. D. 1994, ApJ, 425, 205
van den Bergh, S. 1994, ApJS, 92, 219
Vaughan, T. E., Branch, D., Miller, D.L., & Perlmutter, S. 1995, ApJ, 439, 558
Wang, L., H¨oflich, P., & Wheeler, J. C. 1997, ApJ, 483, L29
Watkins, R., & Feldman, H. A. 1995, ApJ, 453, L73
White, M. 1998, astro-ph/9802295

– 18 –
Figure 1: The solid and dotted lines show the B, V, R, and ICCD transmission functions for the
FLWO 1.2 m telescope as determined from the FLWO passband filters and the quantum efficiency curve of
the FLWO thick and thin CCD, respectively. These are compared to the Bessell (1990) representation of
the Johnson/Cousins convention for the B, V, R, and Iphototube transmissions (dashed line). The sharp
phototube transmission functions of the Johnson/Cousins convention are not perfectly matched using CCD
detectors which are sensitive to light over a range of wavelengths.
Figure 2: Photometry comparison stars in the fields of 22 SNe Ia. The stars used for measuring the
the brightness of each SN Ia are listed in Table 2 and indicated in the figures. The orientation of each field
is North at the top and East to the left. The horizontal arrow at the top left indicates the length of one
arcminute.
Figure 3: B, V , R, and Ilight curves of 22 SNe Ia. The Vlight curves (filled circles) are plotted
without an offset. The Blight curves (open circles) are plotted at ∼+1 mag offset from V, the Rlight
curves (open diamonds) are plotted at ∼-1 mag offset from V, and the Ilight curves (open squares) are
plotted at ∼-2 mag offset from V. The lines are the MLCS empirical fits to the data (Riess, Press, &
Kirshner 1996a; Riess et al. 1998b).
Figure 4: Characteristics of CfA SN Ia sample. Shown with solid lines are histograms of the redshifts
(z), apparent magnitude (mV), epoch of first Vobservation relative to Bmaximum, number of observations,
absolute magnitude (MV) as determined from the luminosity/light-curve parameter, and line-of-sight
visual extinction, AV. The last three parameters are derived from MLCS empirical fits to the data (Riess,
Press, & Kirshner 1996a; Riess et al. 1998b). Shown in dash-dot lines are the same characteristics for the
Cal´an/Tololo Supernova Survey (C/T).
Figure 5: SN Ia extinction and light curve parameter trends with galactocentric distance and host
galaxy morphology. A general decrease of host galaxy extinction is seen with galactocentric distance and
an increase with the lateness of host galaxy type. A weak trend of the MLCS light-curve parameter, ∆, is
that SNe Ia with comparatively faster (dimmer) light curves occur further from the center of galaxies and
are more common to early-type galaxies. The CfA SNe Ia are shown as filled symbols, the C/T SNe Ia are
shown as open symbols, and SNe Ia calibrated by Cepheid variables are shown as X’s.
Figure 6: SN Ia luminosity versus galactocentric distance. The luminosities of SNe Ia, uncorrected

– 19 –
for light-curve shape or extinction (a), display a greater variation closer to the galaxy centers as noted by
Wang, H¨oflich, & Wheeler (1997). After correction for extinction (b), SNe Ia with projected separations of
less than 10 kpc are, on average, brighter by about 0.3 mag than those further out. This relation is also
shown by host galaxy type (d). After the luminosities are corrected for light-curve shape and extinction, no
significant trend with galactocentric distance is apparent (c). The CfA SNe Ia are shown as filled symbols,
the C/T SNe Ia are shown as open symbols.
Figure 7: Hubble diagram of CfA sample 17 SNe Ia with cz > 2500 km s−1. The distances are
determined by empirical MLCS fits to the light curves described by Riess, Press, & Kirshner (1996a), and
updated by Riess et al. (1998b).
Figure 8: Composite B, V , R, and ISN Ia light curves. These light curves were made by normalizing
the 22 CfA SN Ia sample light curves in time and brightness at the fit to the initial peak, including a
correction for 1 + ztime dilation and a K-correction. They include over 1200 individual data points. The
inhomogeneity of the light curves is readily apparent.

4000 6000 8000 10000
Angstroms
0.0
0.2
0.4
0.6
0.8
1.0
1.2
transmission
B V R I FLWO(thick)
FLWO(thin)
Bessell
(1990)

This figure "ariess.fig2.1.gif" is available in "gif" format from:
http://arXiv.org/ps/astro-ph/9810291v1

This figure "ariess.fig2.2.gif" is available in "gif" format from:
http://arXiv.org/ps/astro-ph/9810291v1

10 20 30 40 50 60
Age (days)
22
20
18
16
Magnitude
1993ac
20 40 60 80
Age (days)
20
18
16
14
Magnitude
1993ae
10 20 30 40 50 60 70
Age (days)
21
20
19
18
17
16
15
Magnitude
1994Q
0 20 40 60
Age (days)
20
18
16
14
Magnitude
1994M
0 10 20 30 40
Age (days)
18
17
16
15
14
13
Magnitude
1994S
0 10 20 30
Age (days)
21
20
19
18
17
16
15
Magnitude
1994T
0 20 40 60 80
Age (days)
18
16
14
12
Magnitude
1995D
0 20 40 60 80
Age (days)
20
18
16
14
Magnitude
1995E
0 20 40 60 80
Age (days)
18
16
14
12
Magnitude
1994ae

0 20 40 60 80
Age (days)
18
16
14
12
Magnitude
1995al
0 20 40 60
Age (days)
21
20
19
18
17
16
15
Magnitude
1995ac
0 20 40 60
Age (days)
20
18
16
14
Magnitude
1995ak
0 20 40 60 80
Age (days)
20
18
16
14
Magnitude
1995bd
0 20 40 60 80
Age (days)
20
18
16
Magnitude
1996C
0 20 40 60
Age (days)
16
14
12
Magnitude
1996X
10 20 30 40 50 60 70
Age (days)
18
17
16
15
14
13
12
Magnitude
1996Z
0 10 20 30 40 50 60
Age (days)
24
22
20
18
Magnitude
1996ab
0 10 20 30 40
Age (days)
19
18
17
16
15
14
13
Magnitude
1996ai

0 10 20 30 40 50
Age (days)
21
20
19
18
17
16
15
Magnitude
1996bl
20 40 60 80
Age (days)
20
18
16
14
Magnitude
1996bv
10 20 30 40 50 60
Age (days)
18
16
14
12
Magnitude
1996bk
0 10 20 30
Age (days)
18
17
16
15
14
Magnitude
1996bo

0.00 0.03 0.06 0.09 0.12 0.15
z
0
2
4
6
8
10
# SNe
CfA
C/T
12 15 18 21
mV
0
2
4
6
8
10
# SNe
-15 -10 -5 0 5 10 15
1st Obs. (day)
0
1
2
3
4
5
6
7
# SNe
15 46 78 109 140
# obs
0
2
4
6
8
10
12
# SNe
-20.0 -19.6 -19.2 -18.8
MV (mag)
0
2
4
6
8
# SNe
01234
AV fit (mag)
0
5
10
15
20
25
# SNe

34 36 38
(m-M)0 (mag)
3.0
3.5
4.0
4.5
5.0
log cz

4
3
2
1
0
B mag
4
3
2
1
0
V mag
-20 0 20 40 60 80 100
Age(days)
4
3
2
1
0
R mag
-20 0 20 40 60 80 100
Age(days)
4
3
2
1
0
I mag